With the widespread application of high-voltage power transmission and transforma-tion equipment in China,the problem of induced electricity is becoming increasingly prominent.To ad-dress the difficulties of practical measurement,this paper employs a backpropagation(BP)neural net-work to learn and analyze the data generated by Maxwell simulation software for 500kV high-voltage transmission lines.In Maxwell,the interaction between the induced electricity from transmission lines and biological bodies was analyzed.In the experiments,by varying the distance between a metal plate and the human body as well as the size of the metal plate,the induced voltage and the average current density in the human body were obtained for different configurations.Subsequently,a BP neural net-work in Matlab was used to train 1200 sets of data(70％training set,15％validation set,15％test set),achieving a goodness of fit of 0.99,indicating excellent training performance.Comparative analy-sis shows that the network training results are consistent with the simulation data,effectively confir-ming the model's reliability in predicting induced electricity under different environmental conditions.
维普
万方数据